Most scleractinian corals and octocorals host a single symbiotic zooxanthella clade

The possibility that scleractinian corals and octocorals could change their symbiotic zooxanthellae in response to global warming is alluring. In Goulet (2006; Mar Ecol Prog Ser 321:1–7) I concluded that corals capable of hosting multiple zooxanthella (Symbiodinium) clades may change zooxanthellae, while those hosting one clade may not. Since the majority of corals (77%) host a single Symbiodinium clade, their survival will depend on whether or not the existing symbioses withstand the changing environment. Baker & Romanski (2007; Mar Ecol Prog Ser 335:237–242, this volume) question whether these conclusions apply to scleractinian corals. They subscribe to the null hypothesis that most corals host multiple clades and argue that if sampling does not reveal multiple clades in a coral species, this is due to insufficient sample sizes, sampling sites and/or detection techniques. In this Reply Comment, I demonstrate that sample sizes are adequate: of the scleractinian corals capable of hosting multiple clades, 87.5% have been detected with sample sizes ≤5, and 97.7% with sample sizes ≤10. Within scleractinian coral families only a minority (29%) of species host multiple clades, concurring with the overall finding at the species level. The conclusions in Goulet (2006) are valid, and the message that coral reefs may undergo a reduction in biodiversity should be heeded.